Quantification of liver fibrosis via second harmonic imaging of the Glisson's capsule from liver surface

Xu, Shuoyu; Kang, Chiang Huen; Gou, Xiaoli; Qian, Peng; Yan, Jie; Zhuo, Shuangmu; Cheng, Chee Leong; He, Yuting; Kang, Yijun; Xia, Wuzheng; So, Peter T. C.; Welsch, Roy E.; Rajapakse, Jagath C.; Yu, Hanry

doi:10.1002/jbio.201500001

Cited by 48 publications

(55 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Classification accuracy was found to be lower for scoring early stage liver fibrosis (Stage 0–1) compared to advanced fibrosis (Stage 2–4). The accuracy of all classification models (AUC = 0.85–1.00) using only morphological features was also found to be comparable to previously reported models (AUC = 0.80–0.95) 11 , 22 . For MLR and SVM-based algorithms, the AUC was very much lower as the system performance was affected by insignificant features.…”

Section: Resultssupporting

confidence: 85%

“…Total collagen content and previously described collagen features, including collagen fiber morphology 20 , 21 and collagen fiber connectivity-related texture measurements 22 , were investigated. A total of 21 morphological features specific to tubular-shaped objects including length, width, orientation, cross-link spaces and cross-link density, etc.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Deep learning enables automated scoring of liver fibrosis stages

Wang

et al. 2018

Sci Rep

Self Cite

View full text Add to dashboard Cite

Current liver fibrosis scoring by computer-assisted image analytics is not fully automated as it requires manual preprocessing (segmentation and feature extraction) typically based on domain knowledge in liver pathology. Deep learning-based algorithms can potentially classify these images without the need for preprocessing through learning from a large dataset of images. We investigated the performance of classification models built using a deep learning-based algorithm pre-trained using multiple sources of images to score liver fibrosis and compared them against conventional non-deep learning-based algorithms - artificial neural networks (ANN), multinomial logistic regression (MLR), support vector machines (SVM) and random forests (RF). Automated feature classification and fibrosis scoring were achieved by using a transfer learning-based deep learning network, AlexNet-Convolutional Neural Networks (CNN), with balanced area under receiver operating characteristic (AUROC) values of up to 0.85–0.95 versus ANN (AUROC of up to 0.87–1.00), MLR (AUROC of up to 0.73–1.00), SVM (AUROC of up to 0.69–0.99) and RF (AUROC of up to 0.94–0.99). Results indicate that a deep learning-based algorithm with transfer learning enables the construction of a fully automated and accurate prediction model for scoring liver fibrosis stages that is comparable to other conventional non-deep learning-based algorithms that are not fully automated.

show abstract

Section: Resultssupporting

confidence: 85%

Section: Methodsmentioning

confidence: 99%

Deep learning enables automated scoring of liver fibrosis stages

Wang

et al. 2018

Sci Rep

Self Cite

View full text Add to dashboard Cite

show abstract

“…(20) Measurement of CF might help assess the progression of LF. (21) Although the consequence of CF on liver function remains elusive, we assume that CF may contribute to an increase of the stiffness of the liver by forming fibrous scars on the liver surface.…”

Section: Discussionmentioning

confidence: 99%

Development of Capsular Fibrosis Beneath the Liver Surface in Humans and Mice

Balog

Ogawa

et al. 2019

Hepatology

View full text Add to dashboard Cite

Glisson's capsule is the connective tissue present in the portal triad as well as beneath the liver surface. Little is known about how Glisson's capsule changes its structure in capsular fibrosis (CF), which is characterized by fibrogenesis beneath the liver surface. In this study, we found that the human liver surface exhibits multilayered capsular fibroblasts and that the bile duct is present beneath the mesothelium, whereas capsular fibroblasts are scarce and no bile ducts are present beneath the mouse liver surface. Patients with cirrhosis caused by alcohol abuse or hepatitis C virus infection show development of massive CF. To examine the effect of alcohol on CF in mice, we first injected chlorhexidine gluconate (CG) intraperitoneally and then fed alcohol for 1 month. The CG injection induces CF consisting of myofibroblasts beneath the mesothelium. One month after CG injection, the fibrotic area returns to the normal structure. In contrast, additional alcohol feeding sustains the presence of myofibroblasts in CF. Cell lineage tracing revealed that mesothelial cells give rise to myofibroblasts in CF, but these myofibroblasts disappear 1 month after recovery with or without alcohol feeding. Capsular fibroblasts isolated from the mouse liver spontaneously differentiated into myofibroblasts and their differentiation was induced by transforming growth factor beta 1 (TGF‐β1) or acetaldehyde in culture. In alcohol‐fed mice, infiltrating CD11b+Ly‐6CLow/– monocytes had reduced mRNA expression of matrix metalloproteinase 13 and matrix metalloproteinase 9 and increased expression of tissue inhibitor of matrix metalloproteinase 1, Tgfb1, and interleukin‐10 during resolution of CF. Conclusion: The present study revealed that the structure of Glisson's capsule is different between human and mouse livers and that alcohol impairs the resolution of CF by changing the phenotype of Ly‐6CLow/– monocytes.

show abstract

“…Histological methods can distinguish between a fibrous tissue and a normal one, but this analysis takes several days and does not provide information about pathological changes in dynamics. The second harmonic generation (SHG) method makes it possible to qualitatively and quantitatively evaluate the accumulation of collagen in tissue and thus determine the stage of fibrosis [105][106][107].…”

Section: Promising Methods For Assessing the Structure Function Andmentioning

confidence: 99%

“…The SHG system has been tested in the clinic and can become an alternative to the traditional histological analysis, significantly reducing the time needed for clinical diagnosis [110]. Likewise, the SHG data correlate with morphology, structure, and thickness of collagen in Glisson capsules [107].…”

Section: Promising Methods For Assessing the Structure Function Andmentioning

confidence: 99%

Modern Methods for Assessing the Regenerative Potential of the Liver after Partial Hepatectomy (Review)

Rodimova¹,

Kuznetsova²,

Bobrov³

et al. 2019

Sovrem Tehnol Med

View full text Add to dashboard Cite

The review addresses the main methods for assessing the function and regenerative potential of the liver. They include both the traditional methods, commonly used in clinical practice, and the latest promising techniques suitable for the analysis of cellular and tissue pathology and having a proven diagnostic value.It is known that the dynamics of liver regeneration is reflected in the metabolic status of liver cells, their morphology, and the molecular rearrangement. Therefore, by looking at these parameters we will be able to assess the regenerative potential of the liver as a whole.At present, the most promising method is represented by multiphoton microscopy able to generate the second harmonic; there are also such techniques as coherent anti-stokes Raman spectroscopy (CARS), stimulated Raman scattering (SRS) microscopy, and fluorescence lifetime imaging microscopy (FLIM). In addition, a number of options for analyzing metabolic and structural changes are provided by mass spectrometry, in particular time-of-flight secondary ion mass spectrometry (ToF-SIMS). Studies using these methods with in vivo models and with human biopsy samples demonstrate their relevance in biomedical research and in clinical practice alike.

show abstract

Quantification of liver fibrosis via second harmonic imaging of the Glisson's capsule from liver surface

Cited by 48 publications

References 63 publications

Deep learning enables automated scoring of liver fibrosis stages

Deep learning enables automated scoring of liver fibrosis stages

Development of Capsular Fibrosis Beneath the Liver Surface in Humans and Mice

Modern Methods for Assessing the Regenerative Potential of the Liver after Partial Hepatectomy (Review)

Contact Info

Product

Resources

About